Putting the Cart Before the Horse

Danielle Toppin offers suggestions to improve a pilot survey in the Caribbean that fell short of reaching a wide number of people in groups vulnerable to HIV.

In the development of HIV/AIDS policies and intervention programmes, one of the key challenges faced by planners is meeting the needs of "hidden," and by extension vulnerable, groups. In recognition of this, and faced with increases in the spread of HIV infection, alongside a paucity of reliable national data on behavioural trends, the member states of the Organisation of Eastern Caribbean States moved to conduct behavioural research on elusive and vulnerable populations. To this end, technical and financial assistance was secured from a number of agencies, with the Caribbean Epidemiology Centre (CAREC) spearheading the research.

This was the genesis of the Behavioural and HIV Seroprevalence Surveillance Surveys (BSS), which were piloted in two OECS member states—Antigua and Barbuda (A&B) and St. Vincent and the Grenadines (SVG), from November 2006-January 2007. Employing the Respondent Driven Sampling Method (RDS), a technique which has successfully been used in reaching "hidden" populations, the pilot study sought to reach a wide cross section of members of two particularly at risk groups—Men who have Sex with Men (MSM) and Female Sex Workers (FSW). The research took a two-pronged approach, investigating not only the prevalence of HIV, but also the attitudes and behaviours of members within the groups. Ultimately, the pilot study hoped not only to gauge HIV prevalence and risk behaviours among these two groups, but also to create a road map that would guide further efforts to create effective intervention policies and programmes, as well as prevention and care options throughout the OECS region.

However, at the close of the survey period, the pilot study had failed to meet its objective of reaching a wide number of respondents; in the case of A&B, interviews were secured with only 6 MSMs and 11 FSWs—numbers which contrast sharply with the projected sample sizes at the start of the study period (111 MSMs and 218 FSWs). Despite these low numbers however, unofficial reports suggest that a significant percentage of respondents tested positive for HIV, a reality that is increasingly alarming given the inability to reach the expected numbers of group members. It importantly raises the questions of "why did these groups remain inaccessible?" and "what can be done in the future to avoid such shortcomings?"

The issue of culture is one that is unquestionably present in answering the first question. The importation of the RDS methodology was done primarily on the basis of its success in other cultural contexts, without giving the necessary consideration to its applicability in the Caribbean, and more specifically in the sub-group of the OECS. Certain cultural facts were therefore not taken into consideration. The discrimination faced by these groups, which could hinder their willingness to self-identify as a member of these groups; the small size of the islands; the rates charged by FSWs, which often surpassed the monetary incentives offered for participation in the research; the nature of the relationships that exist within the groups … all of these concerns could have been integrated into the research.

The need for cultural consideration is seen in the case of FSWs in the two pilot countries. In Antigua, sex work is indeed present, yet it is reportedly marked by a hostile and unstable climate in which FSWs may not form numerous relationships with other FSWs, who may be seen as competition. Such women would therefore be unable—or unwilling—to provide the wide reach required in the RDS referral system. In the case of SVG, one of the challenges faced was the weak network of FSWs, another cultural consideration that challenged the efficacy of the RDS methodology.

This raises another consideration. Even within the Caribbean it cannot be assumed that we are a one-dimensional people. Cultures vary from island to island, and even within islands. Research of such a delicate nature must take into account the ways in which the cultural climates vary within and across the islands. Such considerations have to take into account issues such as race, class, nation and sexuality, amongst others.

This leads me to suggesting one possible solution to the second question. In most countries, there are local level community workers whose understanding of these groups far surpasses that of researchers who often come from different territories or social backgrounds. Integrating such key persons into the research process could go a far way towards developing research that works the first time around. Value must be given to local knowledge at all stages of the research process. Too much valuable time is wasted when key players are left out. Given the increases in HIV infections and the lack of information about the prevalence of the disease, as well as the behaviours in vulnerable groups … that valuable time is some we just can't afford to waste.